1. Field of the Invention
The present invention broadly relates to spread spectrum modulation, and especially frequency hopping techniques. More particularly, the invention relates to a method for synchronizing the operation of a transmitter and receiver in a frequency hopping communication system.
2. Description of the Related Art
Frequency hopping is often employed in communication systems in order to spread the signal spectrum so as to provide discrimination against energy-limited interference by using cross-correlation or matched-filter detectors. The interference may be natural (impulse noise), inadvertent (as in amateur radio or aircraft channels), or deliberate, where the jammer may transmit continuous or burst continuous wave, swept continuous wave, narrow-band noise, wide-band noise or replica or deception waveforms. Spread-spectrum systems such as frequency hopping occupy a signal bandwidth much larger than the information bandwidth. For example, in a frequency hopping system, if 100 channels are provided, each having a bandwidth of 300 Hz, the signal bandwidth or hop band is 30 KHz.
In connection with electronic counter measures (ECM), jamming techniques may be employed to interfere with or prevent receipt of the transmitted information. One technique of jamming consists of spreading the jam energy over the entire signal bandwidth. However, this results in the reduction of the jammer energy that can be allocated to each channel. Alternatively, the jammer may allocate all of the jam energy to a single channel. In this approach, the jammer becomes much more effective if he can synchronize his random hop generator with that of the transmitter. In order to achieve such synchronization, the jammer must know the hopping code of the transmitter or be prepared to follow the hopping pattern with great agility. The victim receiver must also know when the transmitter commences to send the hop pattern so that it can be synchronized with the pseudorandom hops of the transmitter. In order to achieve this synchronization, a hopped "preamble" is transmitted which is known in every detail to the victim receiver. The preamble is so elaborate in its details that it could not be mistaken by the victim receiver for any other information-bearing signal. In other words, the preamble is an unambiguous time marker which informs the receiver exactly when the message or information is going to commence. This preamble, however, is indistinguishable to the jammer from a typical information-bearing signal.
In the past, acquisition of initial synchronization to a frequency hopped preamble has been accomplished by one of two techniques. The first technique involves progressively slewing the receiver clock and attempting to correlate the local signal with the receiver clock and attempting to correlate the local signal with the received preamble. This slew-and-compare approach requires only a single RF-1F receiver processor, which is shifted in frequency by the local frequency synthesizer. The second approach to achieving synchronization involves convolving the received signal with a stored replica of the preamble (or some sufficiently long portion of it) and continuously testing for a peak of correlation. The convolving technique requires a bank of RF-1F processors, typically as many processors as there are different frequencies or channels involved in the preamble. The slew-and-compare approach requires a great deal longer to resolve a given time uncertainty at the receiver than the preamble convolver technique. For example, if the hop rate is R hops per second, the preamble consists of N pulses, and the time uncertainty at the receiver is T seconds, then the slew-and-compare technique requires a maximum of RT comparisons, each N/R seconds long, or NT seconds to resolve the uncertainty. In contrast, the convolving approach requires a maximum of only T seconds (plus the duration of the preamble itself, which is typically much less than T). For a large N, the contrast between the two techniques becomes quite significant and the choice between the two approaches requires a trade-off of speed versus complexity. Some applications, however, require a minimum of complexity but a faster processing time than is afforded by the slew-and-compare technique.